DOI: 10.1126/science.1236566, (2013);341 Science

et al.Seong A. KangStarvation and RapamycinmTORC1 Phosphorylation Sites Encode Their Sensitivity to

This copy is for your personal, non-commercial use only.

clicking here.colleagues, clients, or customers by , you can order high-quality copies for yourIf you wish to distribute this article to others

  here.following the guidelines

can be obtained byPermission to republish or repurpose articles or portions of articles

  ): July 25, 2013 www.sciencemag.org (this information is current as of

The following resources related to this article are available online at

http://www.sciencemag.org/content/341/6144/1236566.full.htmlversion of this article at:

including high-resolution figures, can be found in the onlineUpdated information and services,

http://www.sciencemag.org/content/suppl/2013/07/24/341.6144.1236566.DC1.html can be found at: Supporting Online Material

http://www.sciencemag.org/content/341/6144/1236566.full.html#relatedfound at:

can berelated to this article A list of selected additional articles on the Science Web sites

http://www.sciencemag.org/content/341/6144/1236566.full.html#ref-list-1, 26 of which can be accessed free:cites 49 articlesThis article

registered trademark of AAAS. is aScience2013 by the American Association for the Advancement of Science; all rights reserved. The title

CopyrightAmerican Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. (print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by theScience

on

July

26,

201

3w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

mTORC1 Phosphorylation Sites Encode Their Sensitivity to Starvation and RapamycinSeong A. Kang, Michael E. Pacold, Christopher L. Cervantes, Daniel Lim, Hua Jane Lou,

Kathleen Ottina, Nathanael S. Gray, Benjamin E. Turk, Michael B. Yaffe, David M. Sabatini*

Introduction: The mechanistic target of rapamycin (mTOR) complex 1 (mTORC1) protein kinase pro-motes cell growth by controlling major anabolic and catabolic processes in response to a variety of environmental and intracellular stimuli and is deregulated in aging and human diseases such as can-cer and diabetes. Rapamycin, an allosteric inhibitor of mTORC1, is used clinically in organ transplanta-tion and the treatment of certain cancers. Exactly how rapamycin perturbs mTORC1 signaling is poorly understood, and it remains unknown why certain mTORC1 phosphorylation sites are sensitive to the drug, whereas others are not. Here, we test the hypothesis that the inherent capacity of a phosphoryla-tion site to serve as an mTORC1 substrate (a property we call substrate quality) is a key determinant of its sensitivity to rapamycin as well as nutrient and growth factor starvation.

Methods: We measured the in vitro kinase activity of mTORC1 towards short synthetic peptides encompassing single mTORC1 phosphorylation sites and refi ned the established mTORC1 phosphory-lation motif. We introduced subtle mutations into bona fi de mTORC1 phosphorylation sites that we found to enhance or reduce their phosphorylation by mTORC1 in vitro and monitored the correspond-ing changes in the sensitivity of these sites to rapamycin treatment within cells. Finally, we assessed whether the modifi cations of the mTORC1 phosphorylation sites also altered their sensitivities to nutri-ent and growth factor starvation.

Results: The response of an mTORC1 phosphorylation site to rapamycin treatment should depend on the balance between the activity of mTORC1 and of the protein phosphatase(s) that dephosphorylates it. We found that the in vitro kinase activity of mTORC1 toward peptides containing established phos-phorylation sites strongly correlates with the resistance of the sites to rapamycin within cells. More-over, the relative affi nities of the mTOR kinase domain for the peptides also correlated with its capacity to phosphorylate them. In addition to a preference for either proline or a nonproline hydrophobic residue in the +1 position, our refi nement of the mTORC1 phosphorylation motif revealed preferences for noncharged residues surrounding the phosphoacceptor site and for serine over threonine as the phosphoacceptor. Utilizing this improved understanding of the sequence motif specifi city of mTORC1, we were able to manipulate mTORC1 activity toward its phosphorylation sites in vitro and alter their sensitivities to rapamycin treatment within cells. mTORC1 phosphorylation sites also varied in their sensitivities to nutrient and growth factor levels and manipulations in substrate quality were suffi cient to alter their responses to nutrient and growth factor starvation.

Discussion: Our fi ndings suggest that the sequence composition of an mTORC1 phosphoryla-tion site, including the presence of serine or threonine as the phosphoacceptor, is one of the key determinants of whether the site is a good or poor mTORC1 substrate within cells. Even though the phosphorylation of mTORC1 sites is subject to varied regulatory mechanisms, differences in substrate quality are one mechanism for allow-ing downstream effectors of mTORC1 to respond differentially to temporal and intensity changes in the levels of nutrients and growth factors or pharmacological inhibitors such as rapamycin. Such differential responses are likely important for mTORC1 to coordinate and appropriately time the myriad processes that make up the vast starvation program it controls. It is likely that the form of hierarchical regulation we describe for mTORC1 substrates also exists in other kinase-driven signaling pathways.

FIGURES IN THE FULL ARTICLE

Fig. 1. Rapamycin differentially inhibits

mTORC1 phosphorylation sites.

Fig. 2. The kinase activity of mTORC1 toward

peptides encompassing its phosphorylation

sites correlates with their resistance to

rapamycin.

Fig. 3. Conservative modifi cations

to mTORC1 phosphorylation sites

are suffi cient to alter their sensitivity

to rapamycin within cells.

Fig. 4. The sequence composition of

mTORC1 phosphorylation sites encodes

their sensitivity to physiological signals

that regulate mTORC1.

SUPPLEMENTARY MATERIALS

Figs. S1 to S13References

mTORC1 activity

et

arts

bu

Sytil

au

q

mTORC1 Phosphorylation sites encode their sen-sitivity to physiological and pharmacological modulators of mTORC1. Substrate quality is an important determinant of how effectively mTORC1 phosphorylates its substrates in the response to both pharmacological and natural regulators of the kinase.

26 JULY 2013 VOL 341 SCIENCE www.sciencemag.org 364

RESEARCH ARTICLE SUMMARY

READ THE FULL ARTICLE ONLINE

http://dx.doi.org/10.1126/science.1236566

Cite this article as S. A. Kang et al., Science 341, 1236566 (2013). DOI: 10.1126/science.1236566

The list of author affi liations is available in the full article online.*Corresponding author. E-mail: sabatini@wi.mit.edu

Published by AAAS

on

July

26,

201

3w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

mTORC1 Phosphorylation SitesEncode Their Sensitivity to Starvationand RapamycinSeong A. Kang,1,2 Michael E. Pacold,1,2,3,4 Christopher L. Cervantes,1,2 Daniel Lim,5Hua Jane Lou,6 Kathleen Ottina,1,2 Nathanael S. Gray,3,7 Benjamin E. Turk,6Michael B. Yaffe,2,5,8 David M. Sabatini1,2,5*

The mechanistic target of rapamycin (mTOR) complex 1 (mTORC1) protein kinase promotes growthand is the target of rapamycin, a clinically useful drug that also prolongs life span in modelorganisms. A persistent mystery is why the phosphorylation of many bona fide mTORC1 substratesis resistant to rapamycin. We find that the in vitro kinase activity of mTORC1 toward peptidesencompassing established phosphorylation sites varies widely and correlates strongly with the resistanceof the sites to rapamycin, as well as to nutrient and growth factor starvation within cells. Slightmodifications of the sites were sufficient to alter mTORC1 activity toward them in vitro and to causeconcomitant changes within cells in their sensitivity to rapamycin and starvation. Thus, the intrinsiccapacity of a phosphorylation site to serve as an mTORC1 substrate, a property we call substratequality, is a major determinant of its sensitivity to modulators of the pathway. Our results reveal amechanism through which mTORC1 effectors can respond differentially to the same signals.

The mechanistic target of rapamycin (mTOR)is a serine-threonine kinase that serves asthe catalytic subunit of two distinct signal-

ing complexes, mTORC1 and mTORC2 [reviewedin (1)]. Both are central regulators of cell growth,and mTORC1 controls key anabolic and catabolicprocesses in response to diverse cues, includingnutrients, energy, and growth factors. mTORC1is commonly deregulated in human diseases,including cancer, and therefore, there are manyefforts to develop drugs that inhibit its kinaseactivity [reviewed in (2, 3)]. The best known suchdrug is rapamycin, which in a complex withthe 12-kD FK506-binding protein (FKBP12),binds near the mTOR kinase domain (4–7) andpartially inhibits its activity (8–11). Rapamycinhas drawn considerable attention in the last fewyears not only because of its accepted clinicaluses in cancer treatment and organ transplanta-tion (1), but also for its capacity to prolong lifespan in multiple model organisms (12–17). Thishas led to great interest in fully understanding

the mTORC1 pathway and exactly how rapa-mycin affects it.

Results and Discussion

Differential Sensitivity of mTORC1Phosphorylation Sites to RapamycinThe effects of rapamycin on mTORC1 are lessstraightforward than initially realized (18–20).The phosphorylation of several sites that arebona fide mTORC1 substrates—including on thekey translational regulator, eukaryotic translationinitiation factor 4E binding protein 1 (4E-BP1)—is largely resistant to rapamycin treatment, whichmay explain the unexpectedly weak efficacy ofthe drug in several early cancer clinical trials [re-viewed in (21)]. The differential effects of rapamy-cin contrast with those of adenosine triphosphate(ATP)–competitive mTORC1 inhibitors, whichblock the phosphorylation of all mTORC1 phos-phorylation sites regardless of their rapamycinsensitivity (18, 19, 22–24). To investigate whyrapamycin differentially inhibits mTORC1 sites,we initially confirmed that this is the case underour experimental conditions in human embryonickidney 293E (HEK-293E) and murine embryonicfibroblast (MEF) cells. Indeed, although Torin1,a specific ATP-competitive inhibitor of mTOR(18, 25), eliminated the phosphorylation of sev-eral well-established mTORC1 sites, rapamycinaffected that of only a subset of sites (Fig. 1, Aand B, and fig. S1).

In Vitro Activity of mTORC1 TowardSubstrate Phosphorylation Sites PredictsTheir Rapamycin SensitivityThe extent of phosphorylation of an mTORC1site within cells should reflect a balance between

the activity of mTORC1 and of the phospha-tases that dephosphorylate the site. Therefore, itis theoretically possible that certain rapamycin-resistant mTORC1 phosphorylation sites arepoorly dephosphorylated so that even the re-duced activity of rapamycin-bound mTORC1is sufficient to keep them phosphorylated. Sucha model predicts that rapamycin-sensitive and-resistant sites would have very different rates ofdephosphorylation. To test this hypothesis, wetreated cells with a dose of Torin1 that com-pletely inhibits mTORC1 (18) and monitoredthe phosphorylation state of several sites overtime. The results were remarkably clear: Withalmost identical kinetics, Torin1 caused the rap-id dephosphorylation of all the sites, irrespectiveof their rapamycin sensitivity (Fig. 1, C and D).Thus, distinct rates of dephosphorylation cannotexplain the differential sensitivity of mTORC1sites to rapamycin.

We considered the alternative possibility thatit is the level of mTORC1 activity toward a partic-ular site that determines whether it is sensitive orresistant to rapamycin within cells. Certain mTORC1phosphorylation sites might be rapamycin-resistantbecause they are very good substrates for mTORC1,so that even the reduced activity of rapamycin-bound mTORC1 would be sufficient to keep themphosphorylated. Relevant to this model, severalmTORC1 substrates (e.g., 4E-BP1 and Grb10)have both rapamycin-sensitive and -resistant sites(Fig. 1B and fig. S2), which indicates that rapa-mycin sensitivity could be encoded at the levelof an individual phosphorylation site rather thana full-length protein. Thus, we measured the invitro kinase activity of mTORC1 toward shortsynthetic peptides encompassing single mTORC1phosphorylation sites rather than toward intactproteins.

The capacity of mTORC1 to phosphorylatethe peptides varied greatly and correlated stronglywith the resistance of the sites to rapamycin with-in cells (Fig. 2, A and B, and fig. S2). For exam-ple, mTORC1 weakly phosphorylated the peptidecontaining the rapamycin-sensitive threonine 389(T389) site of S6K1 but strongly phosphorylatedthe peptides containing the T37 or T46 sites of4E-BP1 or S150 of Grb10, which are rapamycin-resistant. The rapamycin-sensitive sites of 4E-BP1S65 or Grb10 S476 were, like S6K1 T389, weak-ly phosphorylated. We also measured mTORC1activity toward peptides containing sites fromLARP1 and PATL1, which were identified in phos-phoproteomic studies as having a high likelihoodof being phosphorylated by mTORC1 but are ofunknown rapamycin sensitivity (26, 27). TheS766 site of LARP1 turned out to be a good in vitrosubstrate, comparable to S150 of Grb10, where-as S774 was a poor one (Fig. 2C). It was gratify-ing to us that, in human cells, Torin1 inhibited thephosphorylation of both sites, whereas rapamycinonly inhibited that of S774 (Fig. 2D). Thus, thein vitro activity of mTORC1 toward peptides en-compassing mTORC1 phosphorylation sites can

RESEARCHARTICLE

1Whitehead Institute for Biomedical Research, Nine CambridgeCenter, Cambridge, MA 02142, USA. 2Department of Biology,Massachusetts Institute of Technology (MIT), 77 MassachusettsAvenue, Cambridge, MA 02139, USA. 3Department of CancerBiology, Dana-Farber Cancer Institute, 450 Brookline Avenue,Boston, MA 02115, USA. 4Department of Radiation Oncology,Dana-Farber Cancer Institute, 450 Brookline Avenue, Boston,MA 02215, USA. 5David H. Koch Center for Integrative CancerResearch at MIT, 77 Massachusetts Avenue, Cambridge, MA02139, USA. 6Department of Pharmacology, Yale UniversitySchool of Medicine, 333 Cedar Street, New Haven, CT 06520,USA. 7Department of Biological Chemistry and Molecular Phar-macology, Harvard Medical School, 240 Longwood Avenue,Boston, MA 02115, USA. 8Department of Surgery, Beth IsraelDeaconess Hospital, Harvard Medical School, 330 BrooklineAvenue, Boston, MA 02215, USA.

*Corresponding author. E-mail: sabatini@wi.mit.edu

www.sciencemag.org SCIENCE VOL 341 26 JULY 2013 1236566-1

on

July

26,

201

3w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

predict the rapamycin sensitivity of the siteswithin cells.

The rapamycin-FKBP12 complex is reportedto be a partial inhibitor of mTORC1 (8–11), andindeed, it attenuated the activity of mTORC1, tosome extent, toward every peptide examined (Fig.2E, bottom). However, it had proportionally muchgreater effects on sites that are poor substrates,such as T389 of S6K1 and S65 of 4E-BP1 (Fig.2E, top), to the point that these sites were barelyphosphorylated in the presence of the drug.Given the small size of the peptides, the differ-ential activity of mTORC1 toward them likelyreflects differences in how the peptides interactwith the mTOR kinase domain rather than withother regions of mTORC1. Consistent with thisnotion, a truncation mutant of mTOR that re-tains its kinase domain but does not interact withraptor (11, 28, 29), a substrate-binding subunit ofmTORC1 (30, 31), had similar peptide prefer-

ences and selectivity to those of intact mTORC1(compare Fig. 2Awith Fig. 2F; and see fig. S3). Ingeneral, its in vitro activity was more inhibitedby rapamycin than that of intact mTORC1 (com-pare Fig. 2E with Fig. 2G), which may indicatea role for raptor in stabilizing the mTOR domainkinase so that in its absence mTOR is moresensitive to allosteric inhibition by rapamycin.The binding of truncated mTOR to the peptidesstrongly correlated with its capacity to phospho-rylate them (Fig. 2H), which suggested thatmTORC1 activity toward peptide substrates is,at least in part, a consequence of their relativeaffinities for the mTOR kinase domain. Con-sistent with this possibility, rapamycin-sensitivepeptides had higher affinities for substrate (Km)than resistant peptides in steady-state kinetic analy-ses of mTORC1 activity (Fig. 2I). Moreover, rapa-mycin increased Km values for both types ofpeptide substrates but had minor effects on the

catalytic rate (kcat) values (Fig. 2I). We specu-late that when FKBP12-rapamycin binds to aregion adjacent to the mTOR kinase domain,it may induce a conformational change in thecatalytic pocket of mTOR that reduces the ac-cessibility of substrates. Alternatively, FKBP12-rapamycin may physically obstruct the substratebinding site, which would cause an increase insubstrate Km.

Refinement of the mTORC1Phosphorylation MotifTo test if a causal relationship exists between thecapacity of mTORC1 to phosphorylate a sitein vitro and its rapamycin sensitivity within cells,it was first necessary to understand how to mod-ify a site—defined as the phosphoacceptor serine-threonine and four residues on either side—toalter mTORC1 activity toward it. In a positionalscanning peptide library screen, mTORC1 showed

Fig.1.Rapamycin differ-entially inhibits mTORC1phosphorylation sites.(A) Responses of knownmTORC1 phosphorylationsites in HEK-293E cells andmouse embryonic fibro-blasts (p53–/– MEFs) to1-hour treatments with100 nM rapamycin, 250 nMTorin1, or vehicle control.Cells were grown in DMEMwith 10% FBS and anti-biotics. Subsequently, celllysates were analyzed byimmunoblotting for thelevels and phosphorylationstates of the specified pro-teins. (B) Sequence align-ment of known and putativemTORC1 phosphorylationsites. Positions are num-bered relative to the centralphosphoacceptor serineor threonine, and knownrapamycin-resistant sitesare indicated. (C) Dephos-phorylation of mTORC1phosphorylation sites inresponse to Torin1. p53–/–

MEFs were treated with1 mM Torin1 and lysed atthe indicated time points,and lysates were analyzedas in (A). (D) Quantifica-tion by densitometry ofimmunoblots shown in(A) and (C).

B

confirmed mTORC1 sites

putativemTORC1 sites

4E-BP1

4E-BP1

T37/46

S65

S6K1

T389 S6K1

P

P

S150

S476

Grb10

Grb10P

HEK-293E MEFs

++

A

S183 PRAS40

PRAS40

S758 ULK1

ULK1

P

P

rapamycin:Torin1: -

- -- +

+-- -

-

rapamycin resistant

C

T37/46

S65

4E-BP1

S6K1

T389 S6K1P

S150

S476

Grb10

1 µM Torin1

MEFs

PRAS40

ULK1

S758 ULK1P

time (min): 0 2 10 20 30 60

4E-BP1P

S183 PRAS40P

5 15

Grb10P

D

0

20

40

60

80

100

120

-20200 603010

time with 1 µM Torin1 (min)

rela

tive

phos

phor

ylat

ion

(a.u

.)

4E-BP1 (T37/46)

4E-BP1 (S65)S6K1 (T389)

Grb10 (S150)

Grb10 (S476)

PRAS40 (S183)

ULK1 (S758)

0 60100 nM rapa (min)

26 JULY 2013 VOL 341 SCIENCE www.sciencemag.org1236566-2

RESEARCH ARTICLE

on

July

26,

201

3w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

a strong preference at the +1 position for pro-line, hydrophobic (L, V), or aromatic residues(F, W, Y) (Fig. 1B), as well as lesser selectivityat other positions, including glycine at –1 (fig.

S3) (26). The preference for both proline andnonproline hydrophobic residues in the +1 posi-tion is unusual and distinguishes mTORC1 fromother proline-directed kinases, such as cyclin-

dependent kinases (Cdks) and mitogen-activatedprotein kinases (MAPKs). Using the Grb10 S150peptide as a model substrate, we confirmed thesepreferences and also found that the addition of

0

0.2

0.4

0.6

0.8

B

0 100

Relative intensity of in vitro phosphorylation

S6K1 T389

4E-BP1 S65

Grb10 S476

Lipin1 S472

4E-BP1 T37

4E-BP1 T46

Lipin1 S106

Grb10 S150

D

0

20

40

60

80

100

120

Grb

10 S

150

LAR

P1

S76

6LA

RP

1 S

774

Regulationin cells

phos

pho/

tota

l pep

tides

vehicle

Torin1

rapamycin

S766 S774

0

0.2

0.4

0.6

0.8

LARP1 S774

PRAS40 S212, S221

PATL1 S179

PRAS40 S183

LARP1 S766

mTOR S2481

TFEB S142

PATL1 S184

ULK1 S758

established rapamycinsensitive site

established rapamycininsensitive site

putative mTORC1 site -rapamycin sensitivity unknown

activ

ity n

orm

aliz

ed to

G

rb10

S15

0 (a

.u.)

LARP1

C

no p

eptid

e4E

-BP1

T37

S6K1

T38

9G

rb10

S15

0G

rb10

S47

6

FLAG

pull-down: biotinylated peptide

FLAG mTOR 1295-2549 w/ mLST8

E

020406080

100120

020406080

100120

4E-B

P1

T37

4E-B

P1

S65

Grb

10 S

150

Grb

10 S

476

S6K

1 T3

89

mTORC1

mTOR 1295-2549 w/ mLST8

FKBP12-Rapamycin

FKBP12

* *

0

100

0

100

4E-B

P1

T37

4E-B

P1

S65

Grb

10 S

150

Grb

10 S

476

S6K

1 T3

89

* *

activ

ity n

orm

aliz

ed to

G

rb10

S15

0 (a

.u.) 160

140

180

160140

180

A

F

020406080

100120140

0

20

40

60

80

100

120

140

4E-B

P1

T37

4E-B

P1

T46

4E-B

P1

S65

Grb

10 S

150

Grb

10 S

476

Lipi

n1 S

106

Lipi

n1 S

472

S6K

1 T3

89

mTORC1

mTOR 1295-2549 w/ mLST8

activ

ity n

orm

aliz

ed to

G

rb10

S15

0 (a

.u.)

activ

ity n

orm

aliz

ed to

G

rb10

S15

0 (a

.u.)

160

160

*rapamycin resistant

* * * **rapamycin resistant

%ac

tivity

G H

FKBP12-Rapamycin

FKBP12

activ

ity n

orm

aliz

ed to

G

rb10

S15

0 (a

.u.)

%ac

tivity

4E-B

P1

T37

4E-B

P1

T46

4E-B

P1

S65

Grb

10 S

150

Grb

10 S

476

Lipi

n1 S

106

Lipi

n1 S

472

S6K

1 T3

89

* * * **rapamycin resistant

*rapamycin resistant

**

*

I

FKBP12-rapamycin: - + - + - + - + - +

peptideKm (µM)

kcat (min-1)

kcat / Km (µM-1 min-1)

Grb10 S150

Grb10 S476

Grb10 S476 (rapa)

Grb10 S150 (rapa)

S6K1 T389

S6K1 T389 (rapa)

0

28.8 ± 1.8

111.2 ± 19.0

74.9 ± 14.8

177.7 ± 54.4

>> 1000*

>> 1000*

1.8 ± 0.2

4.0 ± 0.4

1.6 ± 0.2

4.8 ± 1.0

-

-

0.06 ± 0.02

0.04 ± 0.01

0.02 ± 0.01

0.03 ± 0.01

-

-

Fig. 2. The kinase activity of mTORC1 toward peptides encompassingits phosphorylation sites correlates with their resistance to rapamycin.(A) In vitro kinase activity of mTORC1 toward a set of short synthetic pep-tides, each containing an established mTORC1 phosphorylation site, wasanalyzed by autoradiography (representative example shown). Phosphoryl-ation levels of the specified peptides were quantified by densitometry. Dataare means T SD (n = 3 to 5). (B) In vitro kinase activity of mTORC1 towardindicated peptide substrates. Results are displayed as the relative kinaseactivity of mTORC1 for each peptide, and the rapamycin sensitivity of eachsite is indicated. The sequences of all indicated phosphorylation sites areavailable in fig. S2. (C) In vitro kinase activity of mTORC1 toward peptidescontaining indicated LARP1 sites was analyzed as in (A). (D) Rapamycinsensitivity of LARP1 phosphorylation sites in cells. FLAG immunoprecipitatesfrom HEK-293E cells stably expressing FLAG-LARP1 and treated with 100 nMrapamycin, 250 nM Torin1, or vehicle control for 1 hour were analyzed bymass spectrometry and phosphorylation ratios determined from chromato-graphic peak intensities. Data are means T SD (n = 3). *P < 0.05 for dif-ferences between treated and nontreated conditions (Mann-Whitney U test). (E)Effects of rapamycin-FKBP12 on mTORC1 activity toward peptide substrates.Experiment was performed and analyzed as in (A), in the presence ofrecombinant FKBP12 or the complex of FKBP12 and 100 nM rapamycin,by using purified mTORC1 (bottom). Activity normalized to mTORC1treated with FKBP12 alone for individual peptide substrates (top). (F) In vitro

kinase activity of the truncation mutant of mTOR toward peptide substrates.Experiment was performed and analyzed as in (A) using the mTOR truncationmutant (amino acid positions 1295 to 2549) in a complex with mLST8. (G)Effects of rapamycin-FKBP12 on the activity of truncated mTOR toward peptidesubstrates. Experiment was performed and analyzed as in (E) by using themTOR truncation mutant. (H) Binding of the peptide substrates to the mTORkinase domain in the absence and presence of FKBP12-rapamcyin. A pull-down assay with streptavidin-agarose was performed from the mixture ofbiotinylated peptides encompassing established mTORC1 phosphorylation sitesand the FLAG-tagged mTOR truncation mutant in the presence of AMP-PNPand analyzed by immunoblotting for the FLAG tag. For pull-down assays withFKBP12-rapamycin, 100 nM rapamycin was preincubated with 50 ng FKBP12for 30 min and added to the assay mixtures. (I) Steady-state kinetic measure-ments of mTORC1 kinase activity toward individual peptides encompassingmTORC1 phosphorylation sites. Km and kcat values of mTORC1 activity towardindicated peptide substrates were determined by measuring the rate ofmTORC1 phosphorylation over a range of peptide substrate concentrations (0,10, 100, 250, 500, and 1000 mM) at a nonlimiting ATP concentration, 500 mM.For kinetic measurements in the presence of FKBP12-rapamycin, 100 nMrapamycin was preincubated with 50 ng FKBP12 for 30 min and added toreaction mixtures. The steady-state kinetic parameters were obtained by fittingthe reaction rates to the Michaelis-Menten equation. *Note: Km exceeds thehighest concentration tested for the S6K1 T389 peptide.

www.sciencemag.org SCIENCE VOL 341 26 JULY 2013 1236566-3

RESEARCH ARTICLE

on

July

26,

201

3w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

hydrophobic or charged residues promoted ordecreased, respectively, mTORC1 activity towardthem (Fig. 3A). In addition, substitution of thephosphoacceptor serine with threonine (Grb10

S150T) strongly reduced mTORC1 activity towardthis peptide (Fig. 3A and fig. S3). The prefer-ences gleaned from modifying the Grb10 S150peptide were applicable to other mTORC1 sub-

strates. For example, elimination of glutamate andlysine from the 4E-BP1 S65 peptide boosted itsphosphorylation by mTORC1 (fig. S4), which ledus to notice that poor mTORC1 phosphorylation

B

Akt

S47

3A

kt S

473

S0T

S6K

1 T3

89

S6K

1 T3

89 T

0S

D

0

20

40

60

80

100

120

WTS6K1

T389SS6K1

norm

aliz

ed in

vitr

o

S6K

1 ac

tivity

(a.

u.)

*

C

0

20

40

60

80

100

120

SG

K1

S42

2G

Grb

10 S

150

S0TWT

0

100

200

300

activ

ity n

orm

aliz

ed to

G

rb10

S15

0 (a

.u.)

activ

ity n

orm

aliz

ed to

G

rb10

S15

0 (a

.u.)

H

*

*

*

020406080

100120140160180

mTO

RC

1 al

one

pept

ide

alon

eW

TP

+1A

G-1

A

G-3

YG

-3F

G-3

LG

-3D

G-3

E

Grb10 S150

AS

0T

activ

ity n

orm

aliz

ed to

G

rb10

S15

0 (a

.u.)

E

FLAG-S6K1

T389 S6K1P

1 µM Torin1

time (min): 0 2 5 10 15 20 30

WTS6K1

T389SS6K1

S6K1/2-/- MEFs

400

S6

FLAG-S6K1

T/S389 S6K1P

T235/236 S6P

WT (T389) T389SFLAG-S6K1:

Rictor

T1135 RictorP

5 nM rapamycin

time (min): 0 2 10 30 605 120

S6K1/2-/- MEFs

WT (S758)FLAG-ULK1: S758T

ULK1-/- MEFs with shULK2

S/T758 ULK1P

FLAG-ULK1

actin

p62

50 nM rapamycin

time (min): 0 15 60 18030 120

50 nM rapamycin

0 15 60 18030 120

FLAG-S6K1

S389 S6K1P

0

60

0

time in 5 nM rapamycin (min)

pS6K

1 (a

.u.) 120

0

60

120

600time in 50 nM

rapamycin (min)

pULK

1 (a

.u.)

120 180

32P-T235/236 S6peptide

10 120

pS6

(a.u

.)pR

icto

r (a

.u.)

0

60

120

p62

(a.u

)

rela

tive

S6K

1ph

osph

oryl

atio

n

0

60

10080

4020

0 10 20 30time in 1 µM Torin1 (min)

5 nM rapamycin

0 2 10 30 605 120

0

60

0

120

20 12040

0

60

0

120

20 12040

SG

K1

S42

2 S

0T

*

T/S389 S6K1P

FLAG-S6K1

rapamycin

rapamycin (nM): 0 0.001 0.01 0.1 1 10

FLAG-S6K1: WT (T389) T389S

S6K1/2-/- MEFs

S6

T235/236 S6P

Rictor

T1135 RictorP

rapamycin

0 0.001 0.01 0.1 1 10

0 10

0 10

0 1010.1

10.1

10.1rapamycin (nM)

F

FLAG-ULK1

S/T758 ULK1P

p62

actin

ULK1/2-/- MEFs

FLAG-ULK1:WT (S758) S758T

rapamycin

rapamycin (nM): 0 0.1 0.5 1 5 10

rapamycin

0 0.1 0.5 1 5 10

0 101rapamycin (nM)

I

0

60

pS6K

1 (a

.u.) 120

pS6

(a.u

.)pR

icto

r (a

.u.)

0

60

120

0

60

120

0

60

120

pULK

1 (a

.u.)

0

60

120

p62

(a.u

)

T-3A

S+3

A S

0T

T-3A

S+3

A

ULK1 S758

120140

rapamycin: +- -+

0 101

0.1

Fig. 3. Conservative modifications to mTORC1 phosphorylation sitesare sufficient to alter their sensitivity to rapamycin within cells. (A)In vitro kinase activity of mTORC1 toward peptides containing indicatedmodifications to the Grb10 S150 site were analyzed by autoradiography(representative example shown). Phosphorylation levels of the specifiedpeptides were quantified by densitometry. Data are means T SD (n = 3 to 5)(*P < 0.05). (B) In vitro kinase activity of mTORC1 toward peptides encom-passing the hydrophobic motif phosphorylation sites of indicated kinases wasanalyzed by autoradiography as in (A). Data are means T SD (*P < 0.05). (C)Time-dependent responses of wild-type and mutant T389S S6K1 to rapamycin.S6K1–/–S6K2–/– MEFs stably expressing wild-type or T389S S6K1 were treatedwith 5 nM rapamycin for up to 2 hours. Cell lysates were analyzed by immu-noblotting for the levels and phosphorylation states of the specified proteins.Phosphorylation levels of the specified proteins were quantified by den-sitometry (graphs). Experiments were performed at least three times, and arepresentative example is shown. (D) Concentration-dependent responsesof wild-type and mutant S6K1 to rapamycin. S6K1–/–S6K2–/– MEFs stablyexpressing FLAG-tagged wild-type or T389S S6K1 were treated with increasingconcentrations of rapamycin for 20 min and analyzed as in (C). (E) Effectsof rapamycin on in vitro kinase activities of wild-type and mutant T389SS6K1. HEK-293T cells expressing FLAG-tagged wild-type or T389S S6K1

were treated with 50 nM rapamycin or vehicle for 15 min, and the recom-binant protein was purified from lysates using FLAG M2 agarose. Sub-sequently, in vitro kinase activity of wild-type or T389S S6K1 toward a S6peptide containing the T235 and T236 phosphorylation sites was analyzedby autoradiography and quantified by densitometry. Data are means T SD(n = 3) (*P < 0.05). (F) S6K1–/–S6K2–/– MEFs stably expressing wild-type orT389S S6K1 were treated with 1 mM Torin1 for indicated time points. Celllysates were analyzed as in (C) (representative example shown). Data aremeans T SD (n = 3). (G) In vitro kinase activity of mTOR toward peptidescontaining indicated modifications to the ULK1 S758 site were analyzed asin (A). The high level of activity of mTORC1 toward the wild-type ULK1peptide reflects the fact that it contains more than one site phosphorylatedby mTORC1. Data are means T SD (n = 3) (*P < 0.05). (H) Time-dependentresponses of wild-type and mutant S758T ULK1 to rapamycin. Experimentwas performed and analyzed as in (C) with ULK1–/– MEFs stably expressing ashort hairpin RNA against endogenous ULK2, as well as FLAG-tagged wild-type or S758T ULK1. (I) Concentration-dependent responses of wild-type andmutant S758T ULK1 to rapamycin. Experiment was performed and analyzedas in (D) for 2 hours with ULK1–/–ULK2–/– MEFs stably FLAG-tagged wild-typeor S758T ULK1. Note: For all peptide sequences, phosphoacceptor sites arein red text and modified residues in yellow highlight.

26 JULY 2013 VOL 341 SCIENCE www.sciencemag.org1236566-4

RESEARCH ARTICLE

on

July

26,

201

3w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

sites tend to have several charged residues(Fig. 1C and fig. S2). Most interestingly, however,the preference for serine over threonine as thephosphoacceptor was even stronger within thecontext of the S6K1 T389 peptide, such thatthe T389S mutant was a much better mTORC1substrate than the wild-type peptide (Fig. 3Band fig. S5). Consistent results were obtainedfor serine-to-threonine changes in the peptidesfor Akt1 S473 or SGK1 S422, which are mTORC2substrates within cells but, in peptide form,are phosphorylated by mTORC1 (Fig. 3B).Thus, analysis of the sequence motif specificityof mTORC1 revealed a simple way to test thehypothesis that we can increase the rapamycinresistance of a site by making it a better sub-strate for mTORC1.

Manipulation Within Cells of theRapamycin Sensitivity of mTORC1Substrate Phosphorylation SitesTo do so we stably expressed wild-type orT389S S6K1 in MEFs lacking S6K1 and S6K2[S6K1–/–S6K2–/– MEFs (32)] and monitored theeffects of various treatment durations or concen-trations of rapamycin on the phosphorylation ofT/S389 with a phosphospecific antibody thatrecognizes either site equally well (fig. S6). Uponrapamycin treatment, S389 S6K1 was dephos-phorylated with slower kinetics and at higherdoses than wild-type S6K1 (Fig. 3, C and D).Moreover, in mutant-expressing cells, the phos-phorylation of S6 and rictor, established S6K1substrates (33–38), was also more resistant torapamycin (Fig. 3, C and D), which is consistentwith the mutant S6K1 retaining more kinase ac-tivity than its wild-type counterpart in rapamycin-treated cells (Fig. 3E). In response to Torin1,wild-type and T389S S6K1 were dephosphoryl-ated with very similar kinetics, which indicatedthat the phosphatases that act on this site wereunaffected by the T389S mutation (Fig. 3F). Theserine mutation did not confer complete resist-ance to rapamycin, perhaps because it does notsufficiently increase the activity of mTORC1toward S6K1. In addition, it is likely that theintrinsic activity of mTORC1 toward a phospho-rylation site is only one of several determinantsof its rapamycin sensitivity. Other properties thatmay have a role include the exact position of thesite on the intact protein substrate, the secondaryinteractions the protein substrate makes with thekinase, and even perhaps its subcellular localiza-tion. Nevertheless, a single conservative changeto an mTORC1 phosphorylation site is sufficientto alter its response to rapamycin and that of thedownstream events the site controls.

We also tested whether making a site a poorermTORC1 substrate increases its sensitivity to ra-pamycin within cells. We used ULK1, an inducerof autophagy that mTORC1 negatively regu-lates, in part by directly phosphorylating it onS758 (39–41). The ULK1 S758 peptide is an ex-ceptionally good substrate in vitro for mTORC1,likely because it contains more than one phos-

phorylation site (Fig. 3G). Still, a S758T muta-tion was sufficient to strongly reduce the activityof mTORC1 toward the peptide (Fig. 3G and fig.S5). Thus, we reconstituted MEFs lacking ULK1and ULK2 (42, 43) with wild-type or the S758TULK1 mutant and examined the extent of S/T758phosphorylation in response to rapamycin, vary-ing either the treatment time or dose of the drug.The phosphospecific S758 antibody recognizesphosphorylation at position 758 when either ser-ine or threonine is the phosphoacceptor (fig.S6). Although the phosphorylation of wild-typeULK1 was, as expected, largely resistant to ra-pamycin (Fig. 1A), that of the mutant was muchmore sensitive to rapamycin (Fig. 3H and I).Moreover, in the mutant-expressing cells, rapa-mycin caused a stronger activation of autopha-gy, as detected by a greater decrease in p62 anda greater accumulation of LC3-II, than in cellswith wild-type ULK1 (Fig. 3, H and I, and figs.S7 and S8) (44). Hence, as with S6K1, a con-servative change to the ULK1 phosphorylationsite is sufficient to alter its sensitivity to rapa-mycin, as well as that of downstream signalingevents, in this case, autophagy induction. Theseresults indicate that the inherent capacity of aphosphorylation site to serve as an mTORC1substrate (its “substrate quality”) affects how itresponds to the partial inhibition of mTORC1caused by rapamycin. The same may be true formTORC2 because phosphorylation of position473 of Akt1 was more sensitive to low doses ofTorin1 when the normal serine was changed tothreonine (fig. S9).

Substrate Quality Is a Determinant of themTORC1-Regulated Starvation ProgramAs rapamycin is a pharmacological regulator ofmTORC1, we wondered whether mTORC1 phos-phorylation sites respond differentially to thephysiological inputs that control mTORC1, suchas nutrients and growth factors. Note that wefound that the same mTORC1 phosphorylationsites that were rapamycin-sensitive were also moresensitive to a partial decrease in the concentra-tion of amino acids in the cell media. For ex-ample, the phosphorylation of T389 S6K1, whichis extremely rapamycin-sensitive, was strongly re-duced when cells were placed in medium with20% of the normal levels of amino acids (Fig. 4Aand fig. S10). In contrast, the same mediumdid not affect the phosphorylation of S150Grb10, which is also resistant to rapamycin,and phosphorylation of this site became par-tially inhibited only when cells were fully de-prived of amino acids (Fig. 4A and fig. S10).We obtained analogous results when we variedthe amount of serum to which the cells were ex-posed or when we varied the duration of completeamino acid starvation (Fig. 4, A and B, and fig.S10). Thus, bona fide mTORC1 substrates varygreatly in their responses to the same mTORC1-regulating signals.

To test whether differences in substrate qual-ity might underlie these differences in sensitivity

to upstream signals, we used MEF lines express-ing the wild-type or the mutant versions of S6K1or ULK1. Phosphorylation of the T389S S6K1mutant, which was partially resistant to rapamy-cin, was also strongly resistant to a reduction inamino acid concentrations, and this resistancewas reflected, as before, in the phosphoryl-ation states of the S6K1 substrates S6 and rictor(Fig. 4C). Similarly, the phosphorylation of theS758T ULK1 mutant, which was sensitive torapamycin, was also sensitive to a reduction inamino acid concentrations and to complete aminoacid starvation, as were amounts of p62 andLC3-II (Fig. 4D and figs. S11 and S12). Forboth kinases, we obtained analogous results whenwe manipulated serum concentrations (fig. S13).Thus, the substrate property that we call substratequality affects how mTORC1 substrates respondto both pharmacological and natural regulatorsof the kinase. Moreover, in a competitive pro-liferation assay in media containing low con-centrations of amino acids, the MEFs expressingT389S S6K1 outcompeted those expressingwild-type S6K1 (Fig. 4E), which indicated thata change in substrate quality can also affect cellbehavior.

We conclude that the sequence compositionof an mTORC1 phosphorylation site, includingthe presence of serine or threonine as the phos-phoacceptor, is one of the key determinants ofwhether the site is a good or poor mTORC1 sub-strate within cells. Even though the phosphoryl-ation of mTORC1 sites is undoubtedly subject tovaried regulatory mechanisms, we propose thatdifferences in substrate quality are one mechanismfor allowing downstream effectors of mTORC1to respond differentially to temporal and inten-sity changes in the levels of nutrients and growthfactors, as well as pharmacological inhibitorssuch as rapamycin (Fig. 4F). Such differential re-sponses are likely important for mTORC1 tocoordinate and appropriately time the myriadprocesses that make up the vast starvation pro-gram it controls. Last, it is likely that the form ofhierarchical regulation we describe for mTORC1substrates also exists in other kinase-driven sig-naling pathways.

Materials and Methods

MaterialsReagents were obtained from the following sources:antibodies to phospho-T389 S6K1, phospho-S235/S236 S6, phospho-T37/T46 4E-BP1, phospho-S654E-BP1, phospho-S70 4E-BP1, phospho-T183PRAS40, phospho-S758 ULK1, phospho-S150Grb10, phospho-S476 Grb10, phospho-S106Lipin1, phosphor-S472 Lipin1, phosphor-S1135rictor, S6K1, 4E-BP1, PRAS40, FLAG, S6, andRictor from Cell Signaling Technology; an anti-body to Grb10 and horseradish peroxidase–labeled antibody against mouse and secondaryantibody against rabbit antibody from Santa CruzBiotechnology; an antibody to p62 from Progen;antibodies to ULK1, FLAG, and b-actin (clone

www.sciencemag.org SCIENCE VOL 341 26 JULY 2013 1236566-5

RESEARCH ARTICLE

on

July

26,

201

3w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

AC-15), FLAG M2 affinity gel, ATP, FKBP12,amino acids, and insulin from Sigma-Aldrich;[g-32P]ATP from Perkin-Elmer; FuGENE 6,PhosSTOP, and Complete Protease Cocktailfrom Roche; rapamycin from LC Laboratories;Dulbecco’s modified Eagle’s medium (DMEM)

from SAFC Biosciences; inactivated fetal calfserum (IFS), fetal bovine serum (FBS), andSimplyBlue Coomassie G from Invitrogen; aminoacid–free Roswell Park Memorial Institute (RPMI)medium from U.S. Biological; Superose 6 10/300GL from GE Healthcare; bicinchoninic acid assay

(BCA assay) reagent, protein G–Sepharose, strep-tavidin agarose, and immobilized glutathione beadsfrom Thermo Scientific; Whatman grade P81 ion-exchange chromatography paper from Fisher Sci-entific; QIAamp DNAMini Kit, QuikChange XLIImutagenesis kit, and XL10-Gold Competent

Fig. 4. The sequence com-position of mTORC1 phos-phorylation sites encodestheir sensitivity to physio-logical signals that regulatemTORC1. (A) Differential re-sponses of established mTORC1phosphorylation sites to par-tial amino acid or serum starva-tion. p53–/– MEFs were placedin media with 100, 20, or 0%of the normal levels of aminoacids or 10, 2, or 0% FBS for30 min. Cell lysates were ana-lyzed by immunoblotting forthe levels and phosphorylationstates of the specified proteins.(B) Differential responses of es-tablished mTORC1 phosphoryl-ation sites to complete aminoacid starvation. p53–/– MEFswere placed in 0% amino acidmedia for the indicated timepoints. Cell lysates were ana-lyzed as in (A). Phosphorylationlevels of the specified proteinswere quantified by densitom-etry (graph). (C) Differentialconcentration-dependent re-sponses of wild-type and mutantT389S S6K1 to partial aminoacid starvation. S6K1–/–S6K2–/–

MEFs stably expressing FLAG-tagged wild-type or T389SS6K1 were placed in media with100, 50, 20, 10, 5, or 0% of thenormal levels of amino acids for20 min. Cell lysates were ana-lyzed as in (A). Phosphorylationlevels of the specified proteinswere quantified by densitom-etry (graph). (D) Differentialconcentration-dependent re-sponses of wild-type and mutantS758T ULK1 to partial aminoacid starvation. Experiment wasperformed and analyzed asin (C) for 2 hours under par-tial amino acid starvation withULK1–/–ULK2–/– MEFs stablyexpressing FLAG-tagged wild-type or S758T ULK1. (E) A con-servative change to the mTORC1phosphorylation site S6K1 T389is sufficient to alter the pro-liferation rate of cells cultured under partial amino acid starvation. S6K1–/–

S6K2–/– MEFs stably expressing barcoded wild-type and T389S S6K1 weremixed in equal number and cultured in either 100% amino acid RPMI with10% FBS and antibiotics or 20% amino acid RPMI with 10% dialyzed FBSand antibiotics. After 32 population doublings, cells were harvested, and

genomic DNA was isolated and analyzed by quantitative real-time PCR.(F) Model for the role of substrate quality in the regulation of mTORC1phosphorylation sites. Substrate quality is an important determinant of howmTORC1 substrates respond to pharmacological and natural regulators ofthe kinase.

A

T37/46

S65

4E-BP1

S6K1

T389 S6K1P

S150

S476

Grb10

Grb10P

PRAS40

ULK1

4E-BP1P

S183 PRAS40P

S758 ULK1P

(100, 20, 0%) (10, 2, 0%)

E

FLAG-ULK1

S/T758 ULK1P

p62

actin

ULK1/2-/- MEFs

FLAG-ULK1: WT (S758) S758Tamino acids

% AA: 100 50 20 10 5 0

amino acids

100 50 20 10 5 0

0

MEFs

amino acids serum

amino acids

% AA: 100 50 20 10 5 0

amino acids

100 50 20 10 5 0

FLAG-S6K1: WT (T389) T389S

S6K1/2-/- MEFs

T/S389 S6K1P

FLAG-S6K1

T235/236 S6P

Rictor

T1135 RictorP

S6

C

0

60

120

pS6K

1 (a

.u.)

0

60

120

pS6

(a.u

.)

0

60

120

pRic

tor

(a.u

.)

100 2040 0 100 010 100 02040amino acids (%)amino acids (%)amino acids (%)

0

60

120

pULK

1 (a

.u.)

0

60

120

p62

(a.u

.)

100 2040 0 100 40amino acids (%)amino acids (%)

20

D

WT S6K1 MEFs

T389S S6K1 MEFs

0

50

100

25

75

doubling: 0 32 32

% AA: - 100 20

*

T37/46

S654E-BP1P

4E-BP1

S6K1

T389 S6K1P

S150

S476

Grb10

Grb10P

0% amino acidsMEFs

PRAS40

ULK1

S183 PRAS40P

S758 ULK1P

0

20

40

60

80

100

120

-20300 60

time in 0% AA (min)

rela

tive

phos

phor

ylat

ion

(a.u

.)

4E-BP1 (T37/46)

4E-BP1 (S65)S6K1 (T389)

Grb10 (S150)

Grb10 (S476)

time (min): 0 5 10 20 30 45 60

PRAS40 (S183)

ULK1 (S758)

B

popu

latio

n pr

opor

tion

(%)

F

mTORC1 activity

etartsb

uS

ytilau

q

26 JULY 2013 VOL 341 SCIENCE www.sciencemag.org1236566-6

RESEARCH ARTICLE

on

July

26,

201

3w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

cells from Stratagene; SYBR Green polymer-ase chain reaction (PCR) Master Mix from ABApplied Biosystems; and SAM2 biotin cap-ture membrane from Promega. Torin1 was pro-vided by Nathanael Gray (Harvard MedicalSchool) (18).

Cell Lines and Tissue CultureHEK-293E and MEFs were cultured in DMEMwith 10% FBS and antibiotics. HEK-293T wascultured in DMEM with 10% IFS and anti-biotics. HEK-293Es were generously providedby John Blenis (Harvard Medical School), p53–/–

MEFs by David Kwiatkowski (Harvard MedicalSchool), S6K1+/+S6K2+/+ and S6K1–/–S6K2–/–

MEFs by Mario Pende (INSERMU845, MedicalSchool, Paris Descartes University), ULK1+/+ andULK1–/– shULK2 MEFs by Reuben Shaw (Salk In-stitute), and ULK1+/+ULK2+/+ and ULK1–/–ULK2–/–

MEFs by Craig Thompson (Memorial Sloan-Kettering Cancer Center).

To generate stable cell lines, mRNA-encodingplasmids were cotransfected with Delta VPR(pLJM60/61 lentivirus) or Gag-pol envelope(pQCXIP/N retrovirus) and cytomegalovirus–vesicular stomatitis virus glycoprotein CMVVSVG) packaging plasmids into actively grow-ing HEK-293T using FuGENE 6 transfectionreagent as previously described (45). Virus par-ticles containing supernatants were collected at48 hours posttransfection and spun in a centri-fuge to eliminate floating cells, and target cells(100,000 to 1,000,000) were infected in the pres-ence of 8 mg/ml polybrene. The cells were givenor split 24 hours postinfection into fresh mediumcontaining, per milliliter, 2 mg puromycin or 1 mgneomycin. mRNA-expressing cells were analyzed2 to 7 days postinfection.

cDNA Manipulations and MutagenesisThe mTOR truncation mutant (1295–2549) andLARP1 cDNAs were amplified by PCR, and theproducts were subcloned into the Sal I and Xho Isites of the FLAG-tagged pQCXIP (puromycin-resistant) retroviral vector for stable expression.The mLST8 cDNA was amplified by PCR, andthe product was subcloned into the Not I andEco RI sites of the pQCXIN (neomycin-resistant)vector for stable expression. The S6K1 and ULK1cDNAs were amplified by PCR, and the productswere subcloned into the Sal I and Not I sites ofthe pLJM60 (puromycin-resistant) or pLJM61(neomycin-resistant) lentiviral vector for stableexpression. The pLJM60 S6K1, pLJM60/61 ULK1,and pRK5 glutathione S-transferase (GST)–taggedmouse Akt1 plasmids were mutagenized withthe QuikChange XLII mutagenesis kit with oligo-nucleotides obtained from Integrated DNA Tech-nologies. The S6K1, ULK1, and Akt1 mutantsused in our experiments were T389S, S758T, andS473T, respectively. For barcoding pLJM60 S6K1constructs, GGATCC (BamH I) and GGTACC(Kpn I) sequences were inserted in front of thestart codons of wild-type and T389S S6K1, re-spectively, by using the QuikChange XLII

mutagenesis kit with oligonucleotides obtainedfrom Integrated DNA Technologies.

Cell Treatments and Lysisand ImmunoprecipitationsFor rapamycin and Torin1 treatments, 70 to 80%confluent cells were treated with dimethyl sulf-oxide (DMSO) or inhibitors as indicated in figurelegends. Amino acids and serum were titrated asindicated in figure legends. Cells rinsed oncewith ice-cold phosphate-buffered saline (PBS)and lysed in ice-cold lysis buffer [50 mM HEPESpH 7.4, 40 mM NaCl, 2 mM EDTA, 1 mM or-thovanadate, 50 mM NaF, 10 mM pyrophosphate,10 mM glycerophosphate, and 1% Triton X-100or 0.3% CHAPS (for immunoprecipitations)] withone PhosSTOP tablet and one tablet of EDTA-freeprotease inhibitors per 25 ml. The soluble fractionsof cell lysates were isolated by centrifugation at13,000g for 10 min. For FLAG immunoprecipi-tations, 50% slurry of FLAG M2 affinity agarosewas added to the lysates, and the mixtures wereincubated with rotation for 2 to 6 hours at 4°C.Immunoprecipitates were washed three timeswith lysis buffer containing 150 mM NaCl. Im-munoprecipitated proteins were denatured bythe addition of sample buffer, boiled for 5 min,resolved by SDS–polyacrylamide gel electro-phoresis (SDS-PAGE), and analyzed by immu-noblotting as previously described (28).

Purifications of mTORC1 and Truncated mTORmTORC1 purification from HEK-293T cells stablyexpressing FLAG-raptor was performed as de-scribed previously (29). Purification of the trun-cated mTOR mutant from HEK-293T cells stablyexpressing FLAG-mTOR (1295–2549) andmLST8 was also performed as described pre-viously without a gel filtration step (29). Puri-fied recombinant proteins were aliquoted andstored at –80°C.

In Vitro Kinase AssaysIndividual peptide substrates [GYXXXX(S/T)XXXXGRRRRR] were synthesized by the MITKoch Institute Biopolymers and Proteomics CoreFacility and purified by reversed-phase high-performance liquid chromatography (HPLC). Invitro kinase activity of mTORC1 or truncated mTORtoward peptides was determined by incubating0.1 to 0.2 mM peptide with ~100 ng mTORC1or ~20 ng truncated mTOR in reaction buffer(25 mM HEPES pH 7.4, 50 mM KCl, 5 mMMgCl2, and 5 mM MnCl2) containing 50 mMcold ATP and 2 to 5 mCi [g-32P]ATP for 20 to30 min at room temperature. Aliquots (3.3 ml) ofeach reaction were spotted onto P81 ion-exchangechromatography paper in triplicates and quenchedin 0.42% H3PO4. Paper was washed 8 to 10 timesin same solution and dried. Resulting radioactivitywas determined by a phosphoimaging device.For kinase assays with rapamycin, 100 nM rapa-mycin was preincubated with 50 ng FKBP12for 30 min and added to reaction mixtures.FKBP12 was added in excess to ensure that most

of rapamycin would be in an FKBP12-rapamcyincomplex.

For S6K1 kinase assays, recombinant S6K1proteins were purified from HEK-293T stably ex-pressing WTor T389S S6K1 using the same meth-od as for truncated mTOR. S6 peptide substrate(AKRRRLSSLRA) was incubated in 20 ml ofreaction mixture consisting of kinase assay buf-fer (25 mM HEPES, pH 7.4, 50 mM KCl, 5 mMMgCl2, and 5 mM MnCl2), recombinant S6K1,50 mMATP, and 2 to 5 mCi [g-32P]ATP for 30 minat room temperature. Aliquots (3.3 ml) of each re-action were spotted onto P81 ion-exchange chro-matography paper in triplicates and quenched in0.42% H3PO4. Paper was washed 8 to 10 timesin same solution and dried. Resulting radioactiv-ity was determined by Phosphoimager.

Pull-Down Assay with Biotinylated PeptidesBiotinylated peptides were dissolved in kinaseassay buffer, and soluble fractions of cell lysateswere collected by centrifugation at 17,000g for10 min. Preincubated mixtures of peptides and50% slurry of streptavidin-agarose were added toFLAG-tagged mTOR (1295–2549) and incubated inthe presence of 500 nM adenylyl-imidodiphosphate(AMP-PNP) for 4 to 12 hours at 4°C. Pull-downmixtures were washed three times with lysis buf-fer containing 150 mMNaCl. Recombinant mTORprotein was denatured by addition of sample buf-fer, boiled for 5 min, resolved by SDS-PAGE,and analyzed by immunoblotting as previouslydescribed (28). For pull-down assays with rapa-mycin, 100 nM rapamycin was preincubated with50 ng FKBP12 for 30 min and added to pull-down mixtures.

Steady-State Kinetic MeasurementsTo determine the kinetic parameters for peptidephosphorylation, assays were conducted in thepresence of 40 nM mTORC1, various concentra-tions of peptide substrates (0, 10, 100, 250, 500,and 1000 mΜ) and an ATP mixture containing500 mM cold ATP (at least 10-fold above Km),and 2 to 5 mCi [g-32P]ATP in a 30-ml reaction mix-ture. The reaction was initiated by the additionof the ATP mixture. After incubation at roomtemperature, aliquots (3 ml) of each reaction werespotted onto P81 ion-exchange chromatographypaper and quenched in 0.42% H3PO4. The paperwas washed 8 to 10 times in same solution anddried. Resulting radioactivity was determined byphosphoimager. For kinetic measurements withrapamycin, 100 nM rapamycin was preincubatedwith 50 ng FKBP12 for 30 min and added toreaction mixtures. The steady-state kinetic param-eters were obtained by fitting the reaction rates tothe Michaelis-Menten equation using GraphPadPrism version 5.0 (GraphPad Inc.)

Mass Spectrometric AnalysesLARP1 phosphorylation sites were identified bymass spectrometric analysis of trypsin-digestedFLAG-LARP1 purified from HEK293T cells sta-bly overexpressing FLAG-LARP1. The amino acid

www.sciencemag.org SCIENCE VOL 341 26 JULY 2013 1236566-7

RESEARCH ARTICLE

on

July

26,

201

3w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

positions of all LARP1 phosphorylation siteswere numbered according to National Center forBiotechnology Information. Label-free quantifi-cation of LARP1 phosphorylation sites was per-formed with BioWorks Rev3.3 software accordingto the methodology previously described (31, 46).

Positional Scanning Peptide Library Screeningand PWM GenerationPositional scanning peptide library (PSPL) screen-ing was performed. The resulting peptides wereanalyzed with the truncated mTOR mutant aspreviously described (47–49).

Phosphopeptide Recognition byPhosphospecific AntibodiesBiotinylated phosphopeptides (1 ml) at the indi-cated concentrations were spotted on a SAM2Biotin Capture Membrane (Promega) and washedthree times in PBST (PBS with Tween-20). Subse-quently, the washed membrane was analyzedby immunoblotting as previously described (28).Phosphopeptide sequences used are as follows:

T389 S6K1: GGYFLGF[pT]YVAPGRRRRRT389S S6K1: GGYFLGF[pS]YVAPGRRRRRS758 ULK1: GGYFTVG[pS]PPSGGRRRRRS758T ULK1: GGYFTVG[pT]PPSGGRRRRR

Competitive Proliferation AssayS6K1–/–S6K2–/– MEFs stably expressing bar-coded wild-type and T389S S6K1 were mixedin equal numbers (100,000) and placed in 10-cmculture dishes. The mixture of cells was culturedin either 100% amino acid RPMI with 10%FBS and antibiotics or 20% amino acid RPMIwith 10% dialyzed FBS and antibiotics. After32 population doublings, cells were harvestedand genomic DNA was isolated using QIAampDNA Mini Kit. The concentration and purity ofDNA were determined by absorbance at 260 to280 nm. Primers for real-time PCR were obtainedfrom Integrated DNATechnologies. Reactions wererun on an Applied Biosystems Prism machineusing SYBR Green Master Mix (Applied Biosys-tems), and the relative abundance of wild-type andT389S S6K1 was calculated. Primer sequencesused to produce bar code–specific amplicons areas follows:

WT S6K1 forward:GTGGTGGTGCGTCGACGGGATWT S6K1 reverse:CACAATGTTCCATGCCAAGTT389S S6K1 forward:GTGGTGGTGCGTCGACGGGTAT389S S6K1 reverse:CACAATGTTCCATGCCAAGT

References and Notes1. M. Laplante, D. M. Sabatini, mTOR signaling in growth

control and disease. Cell 149, 274 (2012). doi: 10.1016/j.cell.2012.03.017; pmid: 22500797

2. D. A. Guertin, D. M. Sabatini, The pharmacology of mTORinhibition. Sci. Signal. 2, pe24 (2009). doi: 10.1126/scisignal.267pe24; pmid: 19383975

3. D. Benjamin, M. Colombi, C. Moroni, M. N. Hall, Rapamycinpasses the torch: A new generation of mTOR inhibitors.

Nat. Rev. Drug Discov. 10, 868 (2011). doi: 10.1038/nrd3531;pmid: 22037041

4. E. J. Brown et al., A mammalian protein targeted by G1-arresting rapamycin-receptor complex. Nature 369, 756(1994). doi: 10.1038/369756a0; pmid: 8008069

5. D. M. Sabatini, H. Erdjument-Bromage, M. Lui, P. Tempst,S. H. Snyder, RAFT1: A mammalian protein that binds toFKBP12 in a rapamycin-dependent fashion and ishomologous to yeast TORs. Cell 78, 35 (1994).doi: 10.1016/0092-8674(94)90570-3; pmid: 7518356

6. C. J. Sabers et al., Isolation of a protein target of theFKBP12-rapamycin complex in mammalian cells. J. Biol.Chem. 270, 815 (1995). doi: 10.1074/jbc.270.2.815;pmid: 7822316

7. J. W. Choi, J. Chen, S. L. Schreiber, J. Clardy, Structure ofthe FKBP12-rapamycin complex interacting with the bindingdomain of human FRAP. Science 273, 239 (1996).doi: 10.1126/science.273.5272.239; pmid: 8662507

8. E. J. Brown et al., Control of p70 S6 kinase by kinase activity ofFRAP in vivo. Nature 378, 644 (1995). doi: 10.1038/378644c0

9. G. J. Brunn et al., Phosphorylation of the translationalrepressor PHAS-I by the mammalian target of rapamycin.Science 277, 99 (1997).

10. P. E. Burnett, R. K. Barrow, N. A. Cohen, S. H. Snyder,D. M. Sabatini, RAFT1 phosphorylation of the translationalregulators p70 S6 kinase and 4E-BP1. Proc. Natl. Acad.Sci. U.S.A. 95, 1432 (1998). doi: 10.1073/pnas.95.4.1432;pmid: 9465032

11. Z. H. Tao, J. Barker, S. D. H. Shi, M. Gehring, S. X. Sun,Steady-state kinetic and inhibition studies of themammalian target of rapamycin (mTOR) kinase domainand mTOR complexes. Biochemistry 49, 8488 (2010).doi: 10.1021/bi100673c; pmid: 20804212

12. D. E. Harrison et al., Rapamycin fed late in life extendslifespan in genetically heterogeneous mice. Nature 460,392 (2009). pmid: 19587680

13. R. A. Miller et al., Rapamycin, but not resveratrol orsimvastatin, extends life span of genetically heterogeneousmice. J. Gerontol. A Biol. Sci. Med. Sci. 66A, 191 (2011).doi: 10.1093/gerona/glq178; pmid: 20974732

14. I. Bjedov et al., Mechanisms of life span extension byrapamycin in the fruit fly Drosophila melanogaster.Cell Metab. 11, 35 (2010). doi: 10.1016/j.cmet.2009.11.010;pmid: 20074526

15. S. Robida-Stubbs et al., TOR signaling and rapamycininfluence longevity by regulating SKN-1/Nrf and DAF-16/FoxO.Cell Metab. 15, 713 (2012). doi: 10.1016/j.cmet.2012.04.007;pmid: 22560223

16. M. Kaeberlein et al., Regulation of yeast replicative life span byTOR and Sch9 in response to nutrients. Science 310, 1193(2005). doi: 10.1126/science.1115535; pmid: 16293764

17. O. Medvedik, D. W. Lamming, K. D. Kim, D. A. Sinclair, MSN2and MSN4 Link Calorie Restriction and TOR to Sirtuin-MediatedLifespan Extension in Saccharomyces cerevisiae. PLoS Biol. 5,e261 (2007). doi: 10.1371/journal.pbio.0050261

18. C. C. Thoreen et al., An ATP-competitive mammaliantarget of rapamycin inhibitor reveals rapamycin-resistantfunctions of mTORC1. J. Biol. Chem. 284, 8023 (2009).doi: 10.1074/jbc.M900301200; pmid: 19150980

19. M. E. Feldman et al., Active-site inhibitors of mTOR targetrapamycin-resistant outputs of mTORC1 and mTORC2. PLoSBiol. 7, e38 (2009). doi: 10.1371/journal.pbio.1000038;pmid: 19209957

20. A. Y. Choo, S. O. Yoon, S. G. Kim, P. P. Roux, J. Blenis,Rapamycin differentially inhibits S6Ks and 4E-BP1 to mediatecell-type-specific repression of mRNA translation. Proc. Natl.Acad. Sci. U.S.A. 105, 17414 (2008). doi: 10.1073/pnas.0809136105; pmid: 18955708

21. S. A. Wander, B. T. Hennessy, J. M. Slingerland, Next-generationmTOR inhibitors in clinical oncology: How pathway complexityinforms therapeutic strategy. J. Clin. Invest. 121, 1231(2011). doi: 10.1172/JCI44145; pmid: 21490404

22. C. M. Chresta et al., AZD8055 is a potent, selective, andorally bioavailable ATP-competitive mammalian target ofrapamycin kinase inhibitor with in vitro and in vivoantitumor activity. Cancer Res. 70, 288 (2010).doi: 10.1158/0008-5472.CAN-09-1751; pmid: 20028854

23. J. M. García-Martínez et al., Ku-0063794 is a specific inhibitorof the mammalian target of rapamycin (mTOR). Biochem. J.421, 29 (2009). doi: 10.1042/BJ20090489; pmid: 19402821

24. K. Yu et al., Biochemical, cellular, and in vivo activity ofnovel ATP-competitive and selective inhibitors of themammalian target of rapamycin. Cancer Res. 69, 6232(2009). doi: 10.1158/0008-5472.CAN-09-0299;pmid: 19584280

25. Q. S. Liu et al., Discovery of 1-(4-(4-propionylpiperazin-1-yl)-3-(trifluoromethyl)phenyl)-9-(quinolin-3-yl)benzo[h][1,6]naphthyridin-2(1H)-one as a highlypotent, selective mammalian target of rapamycin (mTOR)inhibitor for the treatment of cancer. J. Med. Chem. 53,7146 (2010). doi: 10.1021/jm101144f; pmid: 20860370

26. P. P. Hsu et al., The mTOR-regulated phosphoproteomereveals a mechanism of mTORC1-mediated inhibition ofgrowth factor signaling. Science 332, 1317 (2011).

27. Y. H. Yu et al., Phosphoproteomic analysis identifiesGrb10 as an mTORC1 substrate that negativelyregulates insulin signaling. Science 332, 1322 (2011).doi: 10.1126/science.1199484; pmid: 21659605

28. D. H. Kim et al., mTOR interacts with raptor to form anutrient-sensitive complex that signals to the cell growthmachinery. Cell 110, 163 (2002). doi: 10.1016/S0092-8674(02)00808-5; pmid: 12150925

29. C. K. Yip, K. Murata, T. Walz, D. M. Sabatini, S. A. Kang,Structure of the human mTOR complex I and its implicationsfor rapamycin inhibition. Mol. Cell 38, 768 (2010).doi: 10.1016/j.molcel.2010.05.017; pmid: 20542007

30. K. Hara et al., Raptor, a binding partner of target ofrapamycin (TOR), mediates TOR action. Cell 110, 177(2002). doi: 10.1016/S0092-8674(02)00833-4;pmid: 12150926

31. T. R. Peterson et al., mTOR complex 1 regulates lipin1 localization to control the SREBP pathway. Cell 146, 408(2011). doi: 10.1016/j.cell.2011.06.034; pmid: 21816276

32. M. Pende et al., S6K1(-/-)/S6K2(-/-) mice exhibit perinatallethality and rapamycin-sensitive 5′-terminal oligopyrimidinemRNA translation and reveal a mitogen-activated proteinkinase-dependent S6 kinase pathway. Mol. Cell. Biol. 24,3112 (2004). doi: 10.1128/MCB.24.8.3112-3124.2004;pmid: 15060135

33. P. Jenö, L. M. Ballou, I. Novak-Hofer, G. Thomas,Identification and characterization of a mitogen-activatedS6 kinase. Proc. Natl. Acad. Sci. U.S.A. 85, 406 (1988).doi: 10.1073/pnas.85.2.406; pmid: 3257566

34. D. J. Price, R. A. Nemenoff, J. Avruch, Purification of ahepatic S6 kinase from cycloheximide-treated Rats.J. Biol. Chem. 264, 13825 (1989).pmid: 2760046

35. J. Chung, C. J. Kuo, G. R. Crabtree, J. Blenis, Rapamycin-FKBPspecifically blocks growth-dependent activation of andsignaling by the 70 kd S6 protein kinases. Cell 69, 1227(1992). doi: 10.1016/0092-8674(92)90643-Q;pmid: 1377606

36. D. J. Price, J. R. Grove, V. Calvo, J. Avruch, B. E. Bierer,Rapamycin-induced inhibition of the 70-kilodalton S6protein kinase. Science 257, 973 (1992). doi: 10.1126/science.1380182; pmid: 1380182

37. C. C. Dibble, J. M. Asara, B. D. Manning, Characterizationof Rictor phosphorylation sites reveals direct regulation ofmTOR complex 2 by S6K1. Mol. Cell. Biol. 29, 5657(2009). doi: 10.1128/MCB.00735-09; pmid: 19720745

38. D. Boulbes et al., Rictor phosphorylation on the Thr-1135 sitedoes not require mammalian target of rapamycin complex 2.Mol. Cancer Res. 8, 896 (2010). doi: 10.1158/1541-7786.MCR-09-0409; pmid: 20501647

39. J. Kim, M. Kundu, B. Viollet, K. L. Guan, AMPK and mTORregulate autophagy through direct phosphorylation ofUlk1. Nat. Cell Biol. 13, 132 (2011). doi: 10.1038/ncb2152; pmid: 21258367

40. C. H. Jung et al., ULK-Atg13-FIP200 complexes mediatemTOR signaling to the autophagy machinery. Mol. Biol.Cell 20, 1992 (2009). doi: 10.1091/mbc.E08-12-1249;pmid: 19225151

41. N. Hosokawa et al., Nutrient-dependent mTORC1association with the ULK1-Atg13-FIP200 complexrequired for autophagy. Mol. Biol. Cell 20, 1981(2009). doi: 10.1091/mbc.E08-12-1248;pmid: 19211835

42. D. F. Egan et al., Phosphorylation of ULK1 (hATG1) byAMP-activated protein kinase connects energy sensing tomitophagy. Science 331, 456 (2011). doi: 10.1126/science.1196371; pmid: 21205641

26 JULY 2013 VOL 341 SCIENCE www.sciencemag.org1236566-8

RESEARCH ARTICLE

on

July

26,

201

3w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

43. H. Cheong, T. Lindsten, J. M. Wu, C. Lu, C. B. Thompson,Ammonia-induced autophagy is independent of ULK1/ULK2 kinases. Proc. Natl. Acad. Sci. U.S.A. 108, 11121(2011). doi: 10.1073/pnas.1107969108; pmid: 21690395

44. N. Mizushima, B. Levine, A. M. Cuervo, D. J. Klionsky,Autophagy fights disease through cellular self-digestion.Nature 451, 1069 (2008). doi: 10.1038/nature06639;pmid: 18305538

45. S. M. Ali, D. M. Sabatini, Structure of S6 kinase 1 determineswhether raptor-mTOR or rictor-mTOR phosphorylates itshydrophobic motif site. J. Biol. Chem. 280, 19445 (2005).doi: 10.1074/jbc.C500125200; pmid: 15809305

46. M. P. Stokes et al., Profiling of UV-induced ATM/ATR signalingpathways. Proc. Natl. Acad. Sci. U.S.A. 104, 19855 (2007).doi: 10.1073/pnas.0707579104; pmid: 18077418

47. J. Mok et al., Deciphering protein kinase specificitythrough large-scale analysis of yeast phosphorylation sitemotifs. Sci. Signal. 3, ra12 (2010).

48. J. E. Hutti et al., A rapid method for determiningprotein kinase phosphorylation specificity.Nat. Methods 1, 27 (2004). doi: 10.1038/nmeth708;pmid: 15782149

49. J. C. Obenauer, L. C. Cantley, M. B. Yaffe, Scansite 2.0:Proteome-wide prediction of cell signaling interactionsusing short sequence motifs. Nucleic Acids Res. 31, 3635(2003). doi: 10.1093/nar/gkg584; pmid: 12824383

Acknowledgments: We thank members of the Sabatinilaboratory for helpful discussions, especially S. Wang, D. Kim,C. Thoreen, T. Wang, and J. Cantor and E. Spooner for themass spectrometric analysis of samples. We also thankM. Pende for the S6K1/2 null cells and R. Shaw and C. Thompsonfor the ULK1/2 null cells. This work was supported by grantsfrom the NIH (CA103866 and AI047389 to D.M.S.; ES015339,GM59281, and CA112967 to M.B.Y.) and Department ofDefense (W81XWH-07-0448 to D.M.S.); awards from the

W.M. Keck Foundation and the LAM (lymphangioleiomyomatosis)Foundation to D.M.S.; fellowships from the American CancerSociety and the LAM Foundation to S.A.K. and the DamonRunyon Cancer Research Foundation and the Department ofDefense Breast Cancer Research Program to M.E.P. D.M.S. is aninvestigator of the Howard Hughes Medical Institute. Torin1,the inhibitor used here, is part of a Whitehead–Dana-FarberCancer Institute patent application on which S.A.K., N.S.G.,and D.M.S. are inventors. Shared reagents are subject to amaterials transfer agreement.

Supplementary Materialswww.sciencemag.org/content/341/6144/1236566/suppl/DC1Figs. S1 to S13References

14 February 2013; accepted 5 June 201310.1126/science.1236566

www.sciencemag.org SCIENCE VOL 341 26 JULY 2013 1236566-9

RESEARCH ARTICLE

on

July

26,

201

3w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from